Multiply damaged sites (MDSs) consisting of DMA base damage and strand breaks are produced in cells by certain chemotherapies and radiotherapy. MDSs generated by ionizing radiation can consist of 2-6 damages within < 20 bp and are believed to be more lethal than single lesions. We hypothesize that DNA repair enzymes can convert MDSs to double strand breaks (DSBs), but that non-homologous end- joining (NHEJ) in human cells acts to prevent this conversion of repair intermediates into lethal events. We have developed assays where synthetic MDSs are placed within the luciferase coding region of a plasmid. The plasmid is transferred to cells and a loss of reporter activity, destruction of the plasmid or the introduction of deletions indicates DSB formation. We will utilize these assays in bacteria, mouse embryonic fibroblasts (MEFs) and human fibroblasts proficient and deficient in base excision repair (BER) or non-homologous end-joining (NHEJ). Our first goal (Specific Aims 1 and 2) is to identify bacterial and mammalian DNA repair enzymes that can convert MDSs to DSBs. We will use defined MDSs that simulate ionizing radiation damage to assess the ability of the bacterial AP endonucleases, Fpg and Ogg1/Neil1 to convert MDSs to DSBs. MDSs will be examined that contain 2 or 3 lesions. We hypothesize that the bacterial AP endonucleases will be able to cleave closely opposed AP sites, even in situations where there is near-by base damage, making them prime candidates to sensitize tumor cells to cancer treatments. We have determined that unlike E.coli, mammalian cells do not readily convert MDSs to DSBs and hypothesize that NHEJ is the cause of the difference in the biological outcome of MDSs in the two cell types. By using mammalian cells deficient in the NHEJ proteins, and E.coli that are manipulated to express the prokaryote NHEJ system, our second goal (aim 3) is to test whether the NHEJ proteins prevent the conversion of MDSs to DSBs. We will also identify which NHEJ protein is involved in this DSB avoidance mechanism. In summary, this work will provide basic knowledge that is required to design better cancer treatments. We will identify candidate DNA repair enzymes that can be over-expressed in tumor cells using a gene therapy approach to enhance the lethality of ionizing radiation and chemotherapy, and determine whether it is necessary to disable NHEJ to improve the lethality of the gene therapy. [unreadable] [unreadable] [unreadable]